AI Engineer | Generative AI Specialist | ML Engineer
I'm a passionate AI engineer with expertise in building production-grade generative AI applications, leveraging cutting-edge LLMs, RAG systems, and agentic AI frameworks. I specialize in designing scalable AI solutions and contributing to innovative open-source projects.
- 🎓 B.Tech in Artificial Intelligence & Data Science
- 💼 Trainee Engineer (AI & Software Development) at Amantya Technologies
- 🚀 Building enterprise-grade AI applications with LLMs, RAG, and Agentic AI
- 🔧 Expert in FastAPI, Python, and distributed systems
- 🌟 Active open-source contributor (Agenta, MLflow, Dify)
| Domain | Expertise |
|---|---|
| Generative AI | LLM Integration, Prompt Engineering, Fine-tuning |
| RAG Systems | Semantic Search, Vector Databases, Context Retrieval |
| Agentic AI | Agent Design, Tool Integration, Reasoning Loops |
| NLP | Text Processing, Embeddings, NER |
| Computer Vision | Image Classification, Face Recognition, Object Detection |
| Machine Learning | Supervised Learning, Unsupervised Learning, Model Optimization |
🔍 OpsLens - AI-Powered RAG Incident Investigation Assistant
Enterprise-grade incident investigation platform leveraging RAG (Retrieval-Augmented Generation) with Google Gemini, vector search, and JWT authentication for secure incident analysis and resolution.
Tech Stack:
Key Features:
- 🤖 Advanced RAG with semantic search
- 🔐 JWT-based authentication
- 📊 Vector database integration (Qdrant)
- 🚀 FastAPI backend
- 🔗 LLM-powered incident analysis
📦 DevForge-MCP - Production-Ready Model Context Protocol Server
AI-powered MCP server enabling language models to understand and interact with local repositories through intelligent semantic code retrieval, repository indexing, and contextual search capabilities. Includes 44 developer tools, multi-project workspace management, and Git integration.
Tech Stack:
Key Features:
- 🔍 Semantic code retrieval with AI understanding
- 📑 Comprehensive repository indexing
- 🎯 Context-aware code search across 44 developer tools
- 🏗️ Multi-project workspace management
- 🐳 Full Docker containerization
🐋 Baleen Whale Sound Event Detection - GPU-Accelerated Bioacoustics Pipeline
An automated Sound Event Detection (SED) pipeline that identifies, classifies, and temporally localizes baleen whale vocalizations in Southern Ocean hydrophone recordings, built for the Antarctic Blue & Fin Whale Acoustic Library challenge. Uses a CNN-RNN (CRNN) architecture — a 5-layer 2D CNN frontend feeding a bidirectional GRU — over STFT spectrograms, with a Butterworth high-pass filter and dynamic resampling for signal preprocessing.
Tech Stack:
Key Features:
- 🎧 CNN-RNN (CRNN) architecture with frame-level multi-label detection for overlapping calls
- 🔊 10 Hz Butterworth high-pass filtering + dynamic resampling for hydrophone noise handling
- ⚖️ Vectorized balanced negative-clip mining to address class imbalance
- 🍎 Apple Silicon (MPS) GPU acceleration support
- 📊 Greedy IoU-based cross-site evaluation pipeline with Raven-compatible outputs
📈 LeadSense - Multi-Tenant CRM & Sales Pipeline Platform
A production-ready, multi-tenant CRM platform for sales teams covering the full lead lifecycle — from capture to close — with role-based access control, activity tracking, email invitations, audit logging, and a real-time dashboard. Built on Clean Architecture with a .NET 10 Minimal API backend and a React 19 + TypeScript frontend.
Tech Stack:
Key Features:
- 🏢 Full multi-tenant isolation with SuperAdmin platform management
- 👥 Role-based access control (SuperAdmin / TenantAdmin / User)
- 📋 End-to-end lead pipeline: New → Contacted → Qualified → Proposal Sent → Won/Lost
- 🔒 JWT auth, BCrypt hashing, and rate-limited login endpoint
- 🔍 Full audit logging with per-entity trails and email invitations via Resend
Amantya Technologies
- 🤖 Developed and deployed production-grade generative AI applications
- 📱 Built scalable REST APIs using FastAPI and Python
- 🔗 Implemented LLM integration and prompt engineering workflows
- 🧠 Designed RAG systems with vector database integration
- 🏗️ Architected agentic AI solutions for enterprise automation
- 🐳 Containerized applications using Docker
- 🔄 Optimized backend systems for performance and scalability
Actively contributing to leading AI/ML open-source projects with multiple merged contributions:
| Project | Role | Contributions |
|---|---|---|
| Agenta | Contributor | LLM evaluation framework enhancements |
| MLflow | Contributor | ML experiment tracking improvements |
| Dify | Contributor | AI workflow orchestration features |
✅ Multiple contributions merged and deployed in production
🚀 Working On:
- Advanced Generative AI Applications
- LLM Integration & Fine-tuning
- RAG System Architecture
- Agentic AI Development
- FastAPI Backend Optimization
📚 Learning:
- Multi-Agent Systems
- Advanced Prompt Engineering
- LLM Inference Optimization
- Production ML Systems
🔭 Exploring:
- Edge AI and TinyML
- Autonomous AI Agents
- Vector Database Optimization
- Real-time ML Systems
✨ Goals:
- Contribute to more open-source AI projects
- Build production-grade AI solutions
- Advance the state-of-the-art in GenAI
🤝 Let's Connect & Collaborate
I'm always open to:
🔗 Collaborating on innovative AI/ML projects
💬 Discussing generative AI and LLM architectures
🤝 Open-source contributions
📧 Professional opportunities